[Paper Review] PIANO: Proximity-based User Authentication on Voice-Powered Internet-of-Things Devices

Neil Zhenqiang Gong, Altay Ozen+6 more|arXiv (Cornell University)|Apr 11, 2017

User Authentication and Security Systems6 references3 citations

TL;DR

PIANO is a proximity-based user authentication system for voice-powered IoT devices that leverages built-in speakers, microphones, and Bluetooth to estimate physical distance between a user's personal device (vouching device) and an IoT device requiring access (authenticating device). By ensuring the devices are within a user-defined threshold (e.g., 1 meter), PIANO achieves low false rejection rates (2.8%) and near-zero false acceptance rates (0.3%), enabling secure, efficient, and spoofing-resistant access control without biometric or password overhead.

ABSTRACT

Voice is envisioned to be a popular way for humans to interact with Internet-of-Things (IoT) devices. We propose a proximity-based user authentication method (called PIANO) for access control on such voice-powered IoT devices. PIANO leverages the built-in speaker, microphone, and Bluetooth that voice-powered IoT devices often already have. Specifically, we assume that a user carries a personal voice-powered device (e.g., smartphone, smartwatch, or smartglass), which serves as the user's identity. When another voice-powered IoT device of the user requires authentication, PIANO estimates the distance between the two devices by playing and detecting certain acoustic signals; PIANO grants access if the estimated distance is no larger than a user-selected threshold. We implemented a proof-of-concept prototype of PIANO. Through theoretical and empirical evaluations, we find that PIANO is secure, reliable, personalizable, and efficient.

Motivation & Objective

• To address the growing security risk of unauthorized physical access to voice-powered IoT devices that store sensitive personal data.
• To overcome limitations of traditional authentication methods (e.g., passwords, biometrics) on resource-constrained IoT devices lacking keyboards, touchscreens, or sensors.
• To design a secure, efficient, and accurate proximity-based authentication mechanism that leverages existing hardware (speakers, microphones, Bluetooth) on voice-powered IoT devices.
• To prevent spoofing attacks such as voice replay and frequency-based spoofing through a novel acoustic signal validation protocol.
• To evaluate the system’s performance in real-world environments across multiple scenarios (office, home, street, restaurant) with varying user densities.

Proposed method

• PIANO uses a vouching device (e.g., smartphone or smartwatch) that carries the user’s authenticated identity and communicates with an authenticating device (e.g., smart speaker) via Bluetooth pairing.
• The system employs a novel distance estimation protocol called ACTION, which uses randomized acoustic reference signals transmitted over Bluetooth and measured via speaker playback and microphone recording.
• Each device computes the round-trip time of flight (ToF) of the reference signals using cross-correlation between transmitted and received signals to estimate distance.
• A sanity check is performed to detect replay attacks by verifying signal characteristics in recorded signals, preventing spoofing attempts.
• Authentication is granted only if the estimated distance is below a user-defined threshold (e.g., 1 meter), ensuring proximity-based access control.
• The system assumes distance estimation follows a Gaussian distribution with mean equal to the true distance and a standard deviation σd estimated from empirical data.

Experimental results

Research questions

• RQ1 Can a proximity-based authentication system be built using only existing hardware (speakers, microphones, Bluetooth) on voice-powered IoT devices?
• RQ2 How accurately and reliably can acoustic signals be used to estimate distance between two IoT devices in real-world environments?
• RQ3 Can the system effectively prevent spoofing attacks such as voice replay and wideband signal injection?
• RQ4 What are the trade-offs between authentication threshold, false rejection rate (FRR), and false acceptance rate (FAR) in diverse environments?
• RQ5 How efficient is the system in terms of time and energy consumption for practical deployment?

Key findings

• In a shared office environment, PIANO achieves a false rejection rate (FRR) of 2.8% and a false acceptance rate (FAR) of 0.3% when the authentication threshold is set to 1.0 meter.
• FRR decreases significantly with increasing threshold—halving from 5.6% to 2.8% when the threshold increases from 0.5 to 1.0 meter across all scenarios.
• FAR remains consistently low (0.3% to 0.8%) across all environments and thresholds, with FAR dropping to 0% when the real distance exceeds 10 meters (Bluetooth range).
• The system completes one authentication in approximately 3 seconds and consumes only 0.6% of a smartphone’s battery over 100 authentication attempts, indicating high efficiency.
• All 100 trials of guessing-based and all-frequency-based replay attacks were successfully detected and blocked by the sanity check in the ACTION protocol, demonstrating strong resistance to spoofing.
• Empirical data confirms that the average estimated distance closely matches the true distance, validating the assumption of a Gaussian-distributed estimation error with a standard deviation σd estimated from experimental measurements.